Calibration of a continuous hydrologic simulation model in the urban Gowrie Creek catchment in Toowoomba, Australia

Abstract

Study regionToowoomba, Queensland, Australia Study focusIn this study we derive loss model parameters suitable for use in the dynamic loss Australian Representative Basin Model (ARBM) through the calibration of a continuous simulation hydrologic model. We compare the derived parameters to those published in the literature, and our results highlight the need to develop a database of calibrated loss parameters for urban catchments. New hydrological insightsThe development of design storms for flood modelling commonly uses the initial loss/continuous loss model to estimate the conversion of rainfall to runoff. This loss model, when applied to pervious areas, uses parameters that have been calibrated for gauged rural catchments. These same parameters are often applied to the pervious component of ungauged urban catchments with minimal understanding of the resulting impact on runoff. This research uses a continuous simulation modelling approach to calibrate parameters suitable for use in the ARBM loss model built into the hydrological modelling software XPRAFTS. Through a two-stage calibration approach, the model offered a satisfactory fit (Nash Sutcliffe Efficiency > 0.5) for 9 of the 11 selected storm events, with seven events exceeding a Nash Sutcliffe Efficiency of 0.75. Events used in the calibration/validation included peak flows as low as 9 m3/s and as high as 600 m3/s. Developing these loss model parameters offers new insights into the suitability of a dynamic loss model approach in an urban catchment in regional Australia and provides an alternative to the parameters already available in the literature which were found to overestimate the peak flow in frequent events.